Electronic health records (EHR) data provide a cost and time-effective opportunity to conduct cohort studies of the effects of multiple time-point interventions in the diverse patient population found in real-world clinical settings. Because the computational cost of analyzing EHR data at daily (or more granular) scale can be quite high, a pragmatic approach has been to partition the follow-up into coarser intervals of pre-specified length. Current guidelines suggest employing a 'small' interval, but the feasibility and practical impact of this recommendation has not been evaluated and no formal methodology to inform this choice has been developed. We start filling these gaps by leveraging large-scale EHR data from a diabetes study to develop and illustrate a fast and scalable targeted learning approach that allows to follow the current recommendation and study its practical impact on inference. More specifically, we map daily EHR data into four analytic datasets using 90, 30, 15 and 5-day intervals. We apply a semi-parametric and doubly robust estimation approach, the longitudinal TMLE, to estimate the causal effects of four dynamic treatment rules with each dataset, and compare the resulting inferences. To overcome the computational challenges presented by the size of these data, we propose a novel TMLE implementation, the 'long-format TMLE', and rely on the latest advances in scalable data-adaptive machine-learning software, xgboost and h2o, for estimation of the TMLE nuisance parameters.

One of the most challenging problems in kernel online learning is to bound the model size and to promote the model sparsity. Sparse models not only improve computation and memory usage, but also enhance the generalization capacity, a principle that concurs with the law of parsimony. However, inappropriate sparsity modeling may also significantly degrade the performance. In this paper, we propose Approximation Vector Machine (AVM), a model that can simultaneously encourage the sparsity and safeguard its risk in compromising the performance. When an incoming instance arrives, we approximate this instance by one of its neighbors whose distance to it is less than a predefined threshold. Our key intuition is that since the newly seen instance is expressed by its nearby neighbor the optimal performance can be analytically formulated and maintained. We develop theoretical foundations to support this intuition and further establish an analysis to characterize the gap between the approximation and optimal solutions. This gap crucially depends on the frequency of approximation and the predefined threshold. We perform the convergence analysis for a wide spectrum of loss functions including Hinge, smooth Hinge, and Logistic for classification task, and $l_1$, $l_2$, and $\epsilon$-insensitive for regression task. We conducted extensive experiments for classification task in batch and online modes, and regression task in online mode over several benchmark datasets. The results show that our proposed AVM achieved a comparable predictive performance with current state-of-the-art methods while simultaneously achieving significant computational speed-up due to the ability of the proposed AVM in maintaining the model size.

We consider an online version of the robust Principle Component Analysis (PCA), which arises naturally in time-varying source separations such as video foreground-background separation. This paper proposes a compressive online robust PCA with prior information for recursively separating a sequences of frames into sparse and low-rank components from a small set of measurements. In contrast to conventional batch-based PCA, which processes all the frames directly, the proposed method processes measurements taken from each frame. Moreover, this method can efficiently incorporate multiple prior information, namely previous reconstructed frames, to improve the separation and thereafter, update the prior information for the next frame. We utilize multiple prior information by solving $n\text{-}\ell_{1}$ minimization for incorporating the previous sparse components and using incremental singular value decomposition ($\mathrm{SVD}$) for exploiting the previous low-rank components. We also establish theoretical bounds on the number of measurements required to guarantee successful separation under assumptions of static or slowly-changing low-rank components. Using numerical experiments, we evaluate our bounds and the performance of the proposed algorithm. In addition, we apply the proposed algorithm to online video foreground and background separation from compressive measurements. Experimental results show that the proposed method outperforms the existing methods.

He is also the main author of H2O's Deep Learning. Before joining H2O, Arno was a founding Senior MTS at Skytree where he designed and implemented high-performance machine learning algorithms. He has over a decade of experience in HPC with C /MPI and had access to the world's largest supercomputers as a Staff Scientist at SLAC National Accelerator Laboratory where he participated in US DOE scientific computing initiatives and collaborated with CERN on next-generation particle accelerators. Arno holds a PhD and Masters summa cum laude in Physics from ETH Zurich, Switzerland. He has authored dozens of scientific papers and is a sought-after conference speaker.

Since its launch in 1987, the annual Conference on Neural Information Processing Systems (NIPS) has brought together researchers working on neural networks and related fields, but it later diversified to become one of the largest conferences in machine learning. In recent years, the trend towards deep learning has brought the conference closer to its roots. The 2016 program spanned six days (Dec 5 to 10) and included tutorials, oral and poster presentations, workshops, and invited talks on a broad range of research topics. Following their previous Insights post on ICML 2016, Two Sigma researchers Vinod Valsalam and Firdaus Janoos discuss below the notable advances in deep learning, optimization algorithms, Bayesian techniques, and time-series analysis presented at NIPS 2016. With 550 accepted papers and 50 workshops, the number of attendees more than doubled in the past two years (from more than 2500 in 2014 to over 5000 in 2016), demonstrating rapidly growing interest in machine learning and artificial intelligence. That included strong industry participation (Two Sigma was among the more than 60 sponsors), both for recruiting talent as well as for presenting advances in the field. Several interesting invited talks were given by researchers who are established in both academia and industry.

Logistic regression enables the researcher to explicitly parameterize and estimate a theoretical model. The other techniques to varying degrees put more emphasis on purely empirical estimation from the data. Whether this is advantageous or disadvantageous will of course depend on the situation but as we move to an increasingly Big Data world I believe that theory-based research will have an inherent advantage: with thousands of potential explanatory variables available, the potential for noise to overwhelm the signal increases, unless we have theoretical models to suggest which variables are more likely to convey signal. It's similar to how computers can beat human chess (and now Go) players -- but the strongest players in the world are expert human players combined with computers.

Ocado is using a combination of Google's open source TensorFlow machine learning tools and cloud APIs to support internal AI projects. One such initiative focuses on automating management of the deluge of customer service-related emails the company receives. Ocado is also building a computer vision system in an effort to replace barcode scanning in its warehouses. The machine learning project is still in development but Daniel Nelson, head of data at Ocado's technology division, told Computerworld UK that they hope the technology will help both within its warehouse and delivery processes. Read next: Ocado to replace barcode scanning with AI'vision' to speed packing processes Ocado is also using Google's open sourced TensorFlow deep-learning library for everything from routing algorithms for its robots to move around warehouses, to improving its existing features like demand forecasting, which is currently based on decade-old linear regression models, and predictively suggesting items to add to your basket depending on past shopping habits.

The really cool thing from my perspective about the Estimators API is that using it is a very easy way to create distributed TensorFlow models. Many of the TensorFlow samples that you see floating around on the internets are not distributed -- they assume that you will be running the code on a single machine. People start with such code and then are immeasurably saddened to learn that the low-level TensorFlow code doesn't actually work on their complete dataset. They then have to do lots of work to add distributed training code around the original sample, and who wants to edit somebody else's code? So, please, please, please, if you see a TensorFlow sample that doesn't use the Estimators API, ignore it.

We propose a novel adaptive approximation approach for test-time resource-constrained prediction. Given an input instance at test-time, a gating function identifies a prediction model for the input among a collection of models. Our objective is to minimize overall average cost without sacrificing accuracy. We learn gating and prediction models on fully labeled training data by means of a bottom-up strategy. Our novel bottom-up method first trains a high-accuracy complex model. Then a low-complexity gating and prediction model are subsequently learned to adaptively approximate the high-accuracy model in regions where low-cost models are capable of making highly accurate predictions. We pose an empirical loss minimization problem with cost constraints to jointly train gating and prediction models. On a number of benchmark datasets our method outperforms state-of-the-art achieving higher accuracy for the same cost.

Mikolov et al. (2013) introduced the skip-gram formulation for neural word embeddings, wherein one tries to predict the context of a given word. Their negative-sampling algorithm improved the computational feasibility of training the embeddings. Due to their state-of-the-art performance on a number of tasks, there has been much research aimed at better understanding it. Goldberg and Levy (2014) showed that skip-gram with negative-sampling algorithm (SGNS) maximizes a different likelihood than the skip-gram formulation poses and further showed how it is implicitly related to pointwise mutual information (Levy and Goldberg, 2014). We show that SGNS is a weighted logistic PCA, which is a special case of exponential family PCA for the binomial likelihood. Cotterell et al. (2017) showed that the skip-gram formulation can be viewed as exponential family PCA with a multinomial likelihood, but they did not make the connection between the negative-sampling algorithm and the binomial likelihood. Li et al. (2015) showed that SGNS is an explicit matrix factorization related to representation learning, but the matrix factorization objective they found was complicated and they did not find the connection to the binomial distribution or exponential family PCA.